Imaging and Sensing
Imaging and Sensing
This domain tests how optical images become sampled, noisy, calibrated measurements. It covers detectors, camera parameters, resolution, radiometry, active sensing such as LiDAR, and spectroscopic or interferometric sensing.
Core rules
- Sampling must satisfy the Nyquist criterion for the spatial frequencies present; pixel pitch alone does not define resolution.
- Instantaneous field of view per pixel is approximately (p/f) in radians for pixel pitch (p) and focal length (f).
- SNR depends on signal photoelectrons, shot noise, dark current, read noise, and background; binning can raise SNR but lowers spatial resolution.
- Radiometry balances source radiance, collection aperture, transmission, and detector responsivity.
- Active sensing measures time of flight or phase; range resolution improves with bandwidth or modulation frequency, not just power.
Worked example
A camera has 5 µm pixels and a 50 mm lens. Angular sample per pixel is (5\times10^{-6}/0.050=1\times10^{-4}) rad, or about 0.0057°. At 1 km range, the ground sample distance is (1\times10^{-4}\times1000=0.1) m. If the diffraction spot is larger than one pixel, the system is diffraction-limited rather than detector-limited; if the spot is much smaller, pixels oversample the optics.
Traps
Confusing magnification with resolution is common. Higher pixel count does not guarantee finer detail if the optics or atmosphere blur the image. Ignoring fill factor and microlens behavior misstates sensitivity. In range sensing, timing jitter and bandwidth matter as much as received power.
Exam focus
Identify the limiting factor: diffraction, detector sampling, noise, or calibration.